This continuation begins with three key observations: a) Although limited glia-associated Abeta remains for more than a year, most of the injected or infused beta-peptide (Abeta) is rapidly scavenged by microglia/macrophages, a process key for Abeta removal. Abeta deposits are not stable and may be degraded or contribute to soluble Abeta aggregates that are potentially neurotoxic. b) Persistent plaque-like deposits in rats chronically infused with Abeta can be induced by an initial co- injection (i.e. transient presentation) of the potent anti- inflammatory cytokine, TGFbeta1. c) Abeta deposition by itself may not be sufficient to induce the major neurodegeneration found in AD because human and animal models can have large amyloid burdens with no dementia and limited synapse or neuron loss. We hypothesized that inflammatory factors regulate deposition and neurodegeneration and that TGFbeta1's antiinflammatory effect leads to deactivation of microglial phagocytosis of Abeta. We have now shown that TGFbetas1 and 3 also increase microglial associated Abeta (chemotaxis) and extracellular deposits in vivo as well as in an organotypic hippocampal slice culture (OHSC), a useful system for investigating the impact of cytokines on Abeta trafficking, deposition or degradation. We present new data supporting the hypothesis that TGFbetas and other microglial activation immunomodulators regulate Abeta deposition/degradation in vivo using our infusion paradigm. TGFbeta1 and very different antiinflammatory agents including steroids, curcumin and NS-398 (a specific COX II inhibitor) all increased Abeta deposition, but reduced neurotoxicity. In fact, in our Abeta infusion paradigm, increasing Abeta deposition is negatively correlated with Abeta toxicity across a number of very different manipulations consistent with the hypothesis that soluble Abeta or small aggregates available to neurons are more neurotoxic than massed aggregates walled off by glial scar formation. In contrast to TGFbetas1 and 3 brief exposure to TGFbeta2 causes increased region-specific neuronal Abeta immunoreactivity including in the CA1 and entorhinal cortex. Because TGFbeta2 is elevated in AD brains, this observation raises the hypothesis that TGFbeta2 increases trafficking of Abeta to neurons and results in neurotoxicity (Aim 1) and compromised neuronal function with reduced hippocampal LTP and spatial memory (Aim 2). The hypothesis that elevated TGFbeta2 expression is a correlate of neurotoxicity is also tested in Aim 3 which characterizes overall cytokine profiles in humans and animal models with and without major Abeta accumulation and neurotoxicity. In Aim 4 we test the hypothesis that acute TGFbeta2 induces persistent antiinflammatory cytokine profiles and characterize the impact of chronically elevated antiinflammatory cytokines on Abeta deposit degradation and neurotoxicity in vitro. Aim 5 pursues these goals in vivo. This project is designed to better understand the role of NSAIDs and antiinflammatory cytokines, especially TGFbeta2 in regulating Abeta deposition, trafficking, degradation and neurotoxicity. It should be of direct relevance to the prevention and treatment of Alzheimer's disease.